Forgot your password?
typodupeerror
Businesses Patents

OmniPage Maker Nuance Loses Patent Trial Over OCR Tech 56

Posted by timothy
from the not-to-put-too-fine-a-point-on-it dept.
rtobyr writes "The Recorder is reporting that Nuance and partner Mofo (law firm Morrison Foerster) have lost a suit over patent infringement involving Optical Character Recognition against Russian competitor ABBYY Software House: 'Nuance had accused ABBYY Software House of infringing three of its patents and mirroring its packaging. Both companies market software that uses optical character recognition technology, or OCR, to convert scanned images of text so they can be searched and edited digitally. Represented by a team of lawyers from Morrison & Foerster and Wilson Sonsini Goodrich & Rosati, Nuance argued that ABBYY's FineReader was little more than a copy of its signature product OmniPage. The Burlington, Mass.-based company also sued Lexmark International Inc. for its use of ABBYY's products and sought more than $100 million in total damages from the two companies. Nuance did not prevail on any claims in Nuance Communications v. ABBYY Software House, 08-0912. MoFo partner Michael Jacobs, who is co-lead counsel for Nuance with fellow MoFo partner James Bennett, declined to comment.'" Update: 08/27 18:43 GMT by T : Sorry for the paywalled link; here's a better one. Update: 08/28 16:02 GMT by T : rtobyr adds: “Sorry about the paywalled link. They must have paywalled it after I submitted the story. It was not paywalled at the time of submission.”
This discussion has been archived. No new comments can be posted.

OmniPage Maker Nuance Loses Patent Trial Over OCR Tech

Comments Filter:
  • That MoFo (Score:5, Funny)

    by wooferhound (546132) <tim@@@wooferhound...com> on Tuesday August 27, 2013 @11:51AM (#44686789) Homepage
    That Mofo didn't know what he was talking about . . .
    • by Anonymous Coward

      That Mofo didn't know what he was talking about . . .

      Uh oh, here comes another lawsuit...

    • by Anonymous Coward

      That's a Mofo partner to you, mister.

  • by CajunArson (465943) on Tuesday August 27, 2013 @11:53AM (#44686813) Journal

    1. Article is behind a registration paywall, not that any of the editors bothered to proofread or click the link.
    2. The "editors" probably chose this submission for the sole reason that it says "MoFo" ... I have heard that Beavis & Butthead is back on the air so I guess the Slashdot editors are trying to get back to that level of discourse.

    • by iamhassi (659463) on Tuesday August 27, 2013 @11:59AM (#44686881) Journal
      How are we suppose to read the article without paying? slashdot isn't as good as it use to be
      • by Anonymous Coward

        We wait until someone posts the entire text of the article into the comments section and reward them with mod points, obviously. Or we wait until someone posts a link to a much better written article about the same news event to the comments while complaining that their submission was not posted while this crappy one was. Lastly, this being slashdot, everyone should know that you don't read TFA. I really don't much care what the article has to say. I'm just happy to see the decision go against the paten

        • Unfortunately at the moment there are no decent articles that are not behind a damn paywall or I would have already posted a link.

        • by tebee (1280900) on Tuesday August 27, 2013 @02:19PM (#44688757)

          Patent Trial Ends in Total Loss for MoFo Client

          By Julia Love Contact All Articles
          The Recorder

          August 26, 2013

          SAN FRANCISCO — After a two-week trial, Nuance Communications Inc. came up empty Monday when a jury found that a Russian competitor had not infringed any of its patents or trade dress.

          Nuance had accused ABBYY Software House of infringing three of its patents and mirroring its packaging. Both companies market software that uses optical character recognition technology, or OCR, to convert scanned images of text so they can be searched and edited digitally.

          Represented by a team of lawyers from Morrison & Foerster and Wilson Sonsini Goodrich & Rosati, Nuance argued that ABBYY's FineReader was little more than a copy of its signature product OmniPage. The Burlington, Mass.-based company also sued Lexmark International Inc. for its use of ABBYY's products and sought more than $100 million in total damages from the two companies.

          Nuance did not prevail on any claims in Nuance Communications v. ABBYY Software House, 08-0912. MoFo partner Michael Jacobs, who is co-lead counsel for Nuance with fellow MoFo partner James Bennett, declined to comment.

          From his opening statement to his closing, ABBYY's lead lawyer, Gerald Ivey of Finnegan, Henderson, Farabow, Garrett & Dunner, urged the jury to honor the American spirit of competition.

          "That's what [this verdict] does," he said in an interview Monday. "It allows ABBYY to continue to compete fairly and on equal footing with all the other companies that are interested in the OCR technology that ABBYY is a real leader in developing."

          The trial before U.S. District Judge Jeffrey White revolved around Nuance's U.S. Patent No. 6,038,342, which covers a "trainable template" that is updated during the process of converting scanned images into searchable text. The technology was roundly applauded when OmniPage debuted in 1988, Bennett said during his closing argument.

          "It's not often in a patent case where you have the kind of widespread, third-party corroboration of the breakthrough, revolutionary... nature of an invention," Bennett told the jury. "And that's what we have here."

          Bennett took ABBYY to task not only for infringing Nuance's patents but also for eroding the prices his client could charge for its products with deep discounting.

          "OmniPage and Nuance, from the time that ABBYY entered this market, have been targeted," he said.

          But Ivey insisted that the technology underlying ABBYY's products bears little resemblance to its competitor's. In contrast with Nuance's trainable template, ABBYY's program relies on a system of weighted guesses to determine word variance in context, he explained in an interview Monday.

          "That is a very different philosophical and technological approach," he said.

          Nuance also cried foul over ABBYY's packaging, which for a time made use of similar colors and images. During his closing argument, Ivey questioned the distinctiveness of Nuance's package design. He noted that there had been no documented cases of consumers mistaking the two companies' products. . And he took issue with the suggestion that his client was trying to masquerade as another company.

          "ABBYY has proudly displayed its name on its packages since it entered the U.S.," he said in an interview.

          During his closing argument, Ivey recounted ABBYY's beginnings as a startup, a story reminiscent of many Silicon Valley companies, though it unfolded in Moscow. The company's founder and CEO both testified in English, though it is their second language.

          "Jurors had an opportunity to hear from them directly," he said. "I think that mattered."

          • Isn't reproducing an entire news article inviting our own IP troll suit? IANAL, but reproducing (and not just linking, paraphrasing or quoting in part) an entire news article appears to go beyond the fair use doctrine. Or what's to prevent, say, a newspaper from simply copypasting another newspaper's lead story?

      • by Anonymous Coward

        How are we suppose to not read the article without paying? slashdot isn't as good as it use to be

        FTFY

        • OK, now I'm confused. Are we not supposed to read the article and pay? Or are we not supposed to read the article and not pay? And it really muddies the waters when someone posts a huge block of text containing TFA...
          All I'm really sure of is that this is /. and I'm not going to read the article!
          • It's slashdot. You're supposed to complain bitterly about the paywall, but when they EDIT it and post the free link, your supposed to lose interest. As evidence, I present every post before this and many follow... which one is actually about the topic of the article? The fixed free link has been posted now for some time.

            Since I know these fuckers, I'll tell you who Nuance is. These are the people who buy up companies that get their start helping people with disabilities, and convert them to cash cows th

    • Re: (Score:2, Funny)

      by Anonymous Coward

      They figure since nearly no one reads the article anyways, there's not much point in making sure the article is reachable. I'm confident in the Slashdot community to be able to have a long-winded argument regardless of the availability of the article itself.

    • by Anonymous Coward

      Dude, Timmy's like eight years old. A law firm named "MoFo" is the funniest thing he's ever seen since the FartApp.

    • by Alien54 (180860)
      There should be an alternate source for this sort of thing. Too bad Groklaw had to close up shop because of the government being an idiot.
      • by SteffenM (166724)

        Nope. I still get a JS overlay demanding that I register for an account at The Recorder "Or sign in with LinkedIn!" before I get access to the article.

    • by Dragonslicer (991472) on Tuesday August 27, 2013 @02:07PM (#44688633)

      2. The "editors" probably chose this submission for the sole reason that it says "MoFo" ... I have heard that Beavis & Butthead is back on the air so I guess the Slashdot editors are trying to get back to that level of discourse.

      You can't exactly blame the editors for that one. The firm's domain is mofo.com [mofo.com], and their about page [mofo.com] is titled "About MoFo". The firm fully embraces the name.

    • "MoFo", I suspect, is Morrison & Foerster, mofo.com, an actual law firm that deals with intellectual property issues.
  • So that says something. Not to say that you know every technology's owner because so many are invisible (or until you infringe on tthem), but when someone says one product everyone knows about infringes on a product no one knows about, the product no one knows about must not be all that hot afterall.

    • by Alan Shutko (5101) on Tuesday August 27, 2013 @12:10PM (#44687045) Homepage

      Well, I do know Omnipage. It's been on the market for decades, and was acquired by Scansoft and then by Nuance, who are most well-known for their speech recognition technology.

      The software used to be highly rated but fell in popularity over the years.

      • by maroberts (15852)

        The software used to be highly rated but fell in popularity over the years.

        That seems to be the usual point at which software companies turn into trolls.

      • by sjbe (173966) on Tuesday August 27, 2013 @12:44PM (#44687565)

        The software used to be highly rated but fell in popularity over the years.

        I use Omnipage almost daily, mostly for the batch document processing and it's the best OCR software I've run across. (If you know of something better I'd love to hear about it) I use it to batch process work instructions and manufacturing orders so that I can search for them more easily. All I have to do is put a pdf (or other file) in a particular folder and it takes care of the rest. It really does a surprisingly good job of it.

        • Re: (Score:3, Interesting)

          by BaphometAten (799044)
          We use Papervision Enterprise and Paperflow Pro both by Digitech Systems to batch scan and ocr thousands of pages a day. We have been quite happy with it for the 7 years that we have been using it.
        • Re: (Score:3, Informative)

          by fast turtle (1118037)

          ReadIris is far better then OmniPage at OCR as it supports multiple Languages plus can scan PDF's and create them.

          • by sjbe (173966) on Tuesday August 27, 2013 @02:28PM (#44688869)

            ReadIris is far better then OmniPage at OCR as it supports multiple Languages plus can scan PDF's and create them.

            Omnipage 18 (the version I use) can scan PDFs and create them. I do that daily. Did you mean something else?

            Multiple languages is interesting but not very useful to me since I only need English for what I do. I know Omnipage can recognize Asian characters and I'm pretty sure it can handle characters in most languages since they are mostly the same.

        • TypeReader Pro http://www.expervision.com/ocr-software/desktop-ocr-typereader-7 [expervision.com]

          Many years ago, I worked for a gov. subcontractor scanning and converting military documents to electronic form. We tried everything out at the time, and TypeReader Pro was by far the best.

          Now that was years ago...but to date, I've never found anything that worked as good as it did.

    • I hadn't heard of Nuance, but OmniPage has been the cream of the OCR crop for over a decade. I thought it was owned by the Omni Group (who bring us OmniGraffle, OmniFocus, OmniPlan and OmniOutliner), but it appears that's not the case. So the issue appears to be that Nuance doesn't market the company well, not that the product itself is unknown.

      Wikipedia says

      OmniPage is an optical character recognition application available from Nuance Communications.

      OmniPage was one of the first OCR programs to run on personal computers.[1] It was developed in the late 1980s and sold by Caere Corporation, a company headed by Robert Noyce. The original developers were Philip Bernzott, John Dilworth, David George, Bryan Higgins, and Jeremy Knight.[2][3][4] Caere was acquired by ScanSoft in 2000.[5] ScanSoft acquired Nuance Communications in 2005, and took over its name.[6]

      OmniPage supports more than 120 different languages.[7]

      That said, I fail to see how there could be a valid patent dispute... patents still last 20 years, right? 20 years ago was 1993, by which point OmniPage was already a very mature product (they'd been perfecting multilingual OCR on crappy fax-level document scans for 13 years by that point). Any actual novel inventions (software or otherwise) should have already been released to the public. In fact, I believe ABBYY moved from translation services into the OCR realm about the year 2000, when some of the original OCR patents had expired.

      ABBYY was founded in 1989 by David Yang[4] and was named BIT Software before 1997. ABBYY Group headquarters are located in Moscow with representative offices in Germany (Munich), the UK (Theale), the USA (Milpitas, CA), Japan (Tokyo), Taiwan (Taipei), Russia (Moscow), Ukraine (Kiev), Canada (Ontario), Australia (Sydney), and Cyprus.[5] In 2007, a branch specializing in publishing dictionaries, reference books, encyclopedias and guide-books, ABBYY Press, was established.[6] ABBYY also owns ABBYY Language Services, a high-tech translation and localization agency.[7]

      These guys have been squabbling for the past decade, as each encroaches further onto the other's turf.

    • So that says something. Not to say that you know every technology's owner because so many are invisible (or until you infringe on tthem), but when someone says one product everyone knows about infringes on a product no one knows about, the product no one knows about must not be all that hot afterall.

      How about Apple's Siri, heard of that? Nuance powers the speech recognition. They don't have a ton of consumer-facing products, but they are in fact very well-known in the technology industry.

      • by tompaulco (629533)
        They don't have a ton of consumer-facing products, but they are in fact very well-known in the technology industry.
        The text -to-speech on my Android is Nuance as well. Also, heard of Dragon Naturally Speaking? That's Nuance now.
        • They don't have a ton of consumer-facing products, but they are in fact very well-known in the technology industry.

          The text -to-speech on my Android is Nuance as well. Also, heard of Dragon Naturally Speaking? That's Nuance now.

          Most of the call centers with the annoying speech recognition are running off a Nuance engine, too. On the plus side, I have found that some companies will route you to to an agent quicker if you swear at it.

    • by tompaulco (629533)
      Well, I have heard of both, as has just about anybody who works with OCR at all.
      They both have their strengths and weaknesses. Abbyy tends to have better accuracy overall. I think the biggest weaknesses on both of their parts is that they are not putting as much effort into the engines anymore and are putting more effort into ancillary crap that is best left to third party vendors that know what they are doing. In a lot of cases, they are marketing products that directly compete with companies who have lic
      • by Anonymous Coward

        They both have their strengths and weaknesses. Abbyy tends to have better accuracy overall.

        I use ABBYY FineReader on an occasional basis. Compared to other proprietary OCR software, one strength of Abbyy is that it is relatively lightweight and runs fast. It's absolutely not the bloated POS that Nuance Omnipage is. And it doesn't have that MS ribbon-style interface and other UI novelties that substantially deviate from what is "tried and true".

        • by tompaulco (629533)

          They both have their strengths and weaknesses. Abbyy tends to have better accuracy overall.

          I use ABBYY FineReader on an occasional basis. Compared to other proprietary OCR software, one strength of Abbyy is that it is relatively lightweight and runs fast. It's absolutely not the bloated POS that Nuance Omnipage is. And it doesn't have that MS ribbon-style interface and other UI novelties that substantially deviate from what is "tried and true".

          I'll agree with that. The only things I care about are speed and accuracy. I just want an SDK to hit. I don't need a UI. I want this stuff to run lights out. If I need a UI for something, I will build it, thank you very much. Just concentrate on making your core product better.

  • Mofo, huh? Certainly a bit more aggressively named that the local FAPlawfirm - I had to restrain myself from making a comment the first time one of their associates gave me his email address.

    • by sconeu (64226)

      The MoFos represented Novell vs. SCOX.
      I believe they were on the Oracle team in Oracle v. Google, but my memory may be off there.

    • by drinkypoo (153816)

      Mofo, huh? Certainly a bit more aggressively named that the local FAPlawfirm - I had to restrain myself from making a comment the first time one of their associates gave me his email address.

      I just installed some Spanish front wheel bearing kits on my 300SD... the brand on the box is "FAG". I was wondering why the American page I bought them from said they were made by a reputable european company, but didn't tell me which one.

      • I just installed some Spanish front wheel bearing kits on my 300SD... the brand on the box is "FAG"

        Is that the same FAG, a Schaeffler Group company, that makes the vibration measuring tool it calls the FAG Detector III [fag-detectoriii.de]?

  • by Anonymous Coward on Tuesday August 27, 2013 @12:23PM (#44687245)

    I do developer support for software company that specializes in SDKs that includes (among other things) a pluggable OCR module that has a few OCR engine options... (hence, replying anon) and it's been my experience that all OCR vendors are batshite insane when it comes to trying to protect their intellectual property.

    We used to sell ABBYY as one of our engines, but it was such an unmitigated clusterfark to get the licensing working that we ended up dropping them... internally, we still refer to them as "the OCR engine that shall not be named".

    One engine we currently have requires physical dongles for developers and will quite deliberately crash if you attempt to attach a debugger to the process (good luck troubleshooting stuff)

    One or our engineers worked for a month back and forth trying to just get an evaluation license for one OCR engine and in the end, the process was deemed so egregious we stopped selling their product too.

    I really like the Tesseract engine (a Google Code open source project) but it's slower and less accurate than several of the commercial offerings and is missing features that some folks just can't live without.

    I've used OmniPage ... many many years ago, and their OCR engine wasn't bad back in the day - but couldn't comment nowadays.

    • by bill_mcgonigle (4333) * on Tuesday August 27, 2013 @12:27PM (#44687313) Homepage Journal

      I've used OmniPage ... many many years ago, and their OCR engine wasn't bad back in the day - but couldn't comment nowadays.

      It used to be a great seller back in the early 90's when I was working Mac tech support. The odd thing is, I tried it out a few years ago (c. 2008) on a modern Windows machine and it seemed to be just as accurate as when I used it on an SE/30 in '93.

    • by CajunArson (465943) on Tuesday August 27, 2013 @12:33PM (#44687391) Journal

      I've used tesseract + ghoscript as a front end to do OCRs of PDF documents. From my experience, tesseract is OK if you have original images that are pretty high quality (300 DPI minimum) printed using standard fonts with pretty standard layouts (the newest versions mostly works OK with a basic 2 column format). You'll still only get results in the high 90% range (which sounds good but is actually pretty atrocious compared to high-end OCR systems that are well up into the 9's for reliability). Oh, and even though you specify a language, tesseract has very little contextual knowledge of what it is scanning so you'll regularly see it run together two letters in properly spelled words to come up with mispelled words.

      Oh, and you have to have a blacklist of characters since tesseract is absolutely in love with the idea of the letter A with the circle coming out of the top even though you tell tesseract that you are specifically scanning English documents where you just have the plain ordinary letter "A". A few other characters are like that too.

      If, however you leave the reservation of high-quality scans of standard black & white printed text with normal layouts, tesseract quickly turns into a lovely random noise generator.

      • by Bill Dimm (463823)

        tesseract is absolutely in love with the idea of the letter A with the circle coming out of the top even though you tell tesseract that you are specifically scanning English

        That character is used as an abbreviation for "angstroms" in English, which is probably why specifying English doesn't eliminate it. Of course, if the software was smart it would realize that the probability of an A-like character being an angstrom sign rather than an "A" is very small, especially if the preceding character wasn't a numeric digit (possibly with a space in between).

  • Can't tell but it seems this is not about the dancing bear of OCR but rather its use and package layout?

    If so, of even less interest.

  • How is ABBYY formed?

  • If you know anything at all of the sordid history of this company, from its beginnings as a Xerox division, to its spinoff as ScanSoft, to its sneaky assimilation of former biggest competitors and continuing to sell multiple formerly competing products including OmniPage, to its current incarnation as Nuance, this lawsuit would not surprise you but the verdict might. Nuance has been getting its way far too often over much of the last decade, and no doubt expected getting its way with this lawsuit. Bazinga, bastards!

  • They must have paywalled it after I submitted the story. It was not paywalled at the time of submission.

When the weight of the paperwork equals the weight of the plane, the plane will fly. -- Donald Douglas

Working...